Earnings call transcript: Loomis beats Q1 2026 expectations, stock dips

Loomis beat Q1 2026 expectations with EPS of SEK 7.88 versus SEK 6.90 consensus and revenue of SEK 7.5 billion versus SEK 7.42 billion, a 14.2% EPS surprise. EBITDA margin rose to 12.6% and U.S. operating profit exceeded SEK 700 million with a record 17.9% margin, though the stock fell 2.67% as investors focused on broader market and geopolitical risks. Management reiterated 2026 growth plans, highlighted a dividend of SEK 15 plus SEK 5 extra per share, and outlined the Hermes acquisition in Peru.

Analysis

The market is treating this as a one-day “AI tax scare,” but the broader signal is that investors are reflexively de-risking anything linked to data-center capex and semiconductor supply chains when policy uncertainty hits. That creates a temporary valuation disconnect in the semiconductor complex: the fundamental damage from a tax regime change is much slower than the tape suggests, while the immediate impact is multiple compression on anything with the word “AI” in the narrative. In that setup, quality leaders with secular demand and pricing power typically recover first once the policy headline fades.

The more interesting second-order read is on capital allocation, not just semis. If tax policy or tariff rhetoric starts to cloud AI infrastructure ROI, hyperscalers may stretch deployment schedules rather than cancel them, which hurts the entire equipment stack in the near term but ultimately benefits the strongest platform vendors once budgets are reprioritized. That would favor the most embedded names and punish the more cyclical, supply-chain-heavy beneficiaries whose earnings are already being discounted aggressively.

Contrarianly, the selloff may be overdone if the market is conflating a policy headline with an earnings reset. A true demand shock would show up in revisions, lead times, and order cancellations over the next 1-2 quarters; absent that, this is more likely a volatility event than a thesis break. The risk is that the market has been willing to pay a large premium for AI growth, so even modest uncertainty can compress multiples another 10-15% before fundamentals catch up.

Key Decisions for Investors

• Buy NVDA on a 1-2 week dislocation if the stock is still trading below pre-headline levels; target 8-12% upside on mean reversion, with a tight 4-5% stop if policy noise escalates into concrete legislation.
• Pair trade: long NVDA / short a basket of higher-beta semiconductor suppliers for 1-3 months; thesis is that leaders recover faster than supply-chain names if AI capex pauses are only deferred, not canceled.
• If using options, buy 1-2 month NVDA calls financed by selling out-of-the-money puts only if implied vol remains elevated; risk/reward improves if the market is pricing policy risk beyond the actual earnings impact.
• Avoid adding to crowded AI-exposed momentum baskets for now; wait for a second leg of selling or for policy clarification, because the first move is likely sentiment-driven rather than fundamental.
• If the tape stabilizes and AI semis bounce, rotate into the highest-quality semicap names rather than chasing the weakest reflexive names; leadership should reassert within days if this is only headline noise.